Real-world mget fan-out and cluster size?

2011-10-05 Thread Fundou
We are planning to deploy a memcached cluster of 40 odd machines and were wondering about mgets and the size of the cluster. We could either: - have all the machines in one cluster, which would mean that an mget on a set of keys could potentially span the entire cluster - partition our cluster in

Re: Real-world mget fan-out and cluster size?

2011-10-06 Thread dormando
> We are planning to deploy a memcached cluster of 40 odd machines and > were wondering about mgets and the size of the cluster. > > We could either: > - have all the machines in one cluster, which would mean that an mget > on a set of keys could potentially span the entire cluster > - partition o

Re: Real-world mget fan-out and cluster size?

2011-10-06 Thread Fundou
Right. It does sound like we'll have to conduct some experiments. Would've been nice to get some other inputs though. We might make 2000-3000 mgets/sec. 100-500 keys each. On Oct 6, 1:38 pm, dormando wrote: > > We are planning to deploy a memcached cluster  of 40 odd machines and > > were wonder

Re: Real-world mget fan-out and cluster size?

2011-10-06 Thread dormando
> Right. It does sound like we'll have to conduct some experiments. > Would've been nice to get some other inputs though. > > We might make 2000-3000 mgets/sec. 100-500 keys each. > It's straight up math. if (keylength * keycount > tcp_buffer) -> latency benefit from having shorter mgets (or just